AI-Enhanced Breast Cancer Screening Improves Recall Imaging, Diagnosis Rates

New research on out-of-pocket artificial intelligence (AI)-enhanced breast cancer screening efficiency was presented at the Radiological Society of North America annual meeting. The study included 747,604 women who underwent screening mammography over an initial 12-month period across 10 health care practices.

More than one third of the participating women opted to enroll in the self-pay AI screening program. Overall breast cancer detection rate was, on average, 43% higher (5.95 vs. 4.15 per 1000) for women who enrolled in the self-pay AI program than for women who did not enroll. This trend was observed at all 10 practices.

Further analysis revealed that 21% of the increase in cancer detection occurred with the AI program and credited the remaining 22% increase in detection, as those who chose to enroll were at increased risk for breast cancer.

“These data indicate that many women are eager to utilize AI to enhance their screening mammogram, and when AI is coupled with a safeguard review, more cancers are found,” noted senior study author Gregory Sorensen, MD.

The recall rate for additional imaging was 21% higher for enrolled women than for unenrolled women. The positive predictive value for cancer was 15% higher for those enrolled, indicating that recall imaging resulted in more cancer diagnoses.

Reference

Sorensen G, Haslam B, Louis L, Holt JS, Storella JM. Women pay for AI to boost mammogram findings. Presented at Radiological Society of North America 2024. December 2-5, 2024. Chicago, Illinois.